Mosquito repellents are used worldwide to prevent mosquito bites and mosquito-borne diseases. In order to develop more efficient and safer mosquito repellents, we first need to understand how currently used repellents work, or how they affect the mosquito’s sense of smell. At the Potter lab, we developed a method (calcium imaging) to visualize how sensory neurons on the antennae of the malaria mosquito respond to repellents. We found that natural repellents, such as lemongrass oil and eugenol, are able to activate some of these neurons. To our surprise, man-made repellents such as DEET did not activate any of these antennal neurons. This suggests that the malaria mosquito can smell natural repellents but not man-made repellents. We found that man-made repellents work in another way: They make neurons on the mosquito antennae respond less (or not at all) to the odors on our skin by decreasing the volatility of these odors. This means that instead of directly repelling the malaria mosquitoes away, man-made repellents prevent chemicals on our skin from evaporating and reaching the mosquitoes, rendering us invisible to them. We further showed that other species of mosquitoes, such as the yellow fever mosquito and the southern house mosquito, are directly repelled by DEET. This suggests that confusion in the field about how DEET works could be due to its species-specific effect. Our findings can help inform mosquito repellent choice by species, and streamline the discovery of improved insect repellents.

As a postdoctoral fellow in the lab of Jeffrey Rothstein, I am studying how alterations in the composition and function of the nuclear pore complex (NPC) contribute to neurodegenerative disease pathogenesis, specifically in Amyotrophic Lateral Sclerosis (ALS) and Frontotemporal Dementia (FTD). The NPC critically governs multiple cellular processes including gene expression and the transport of macromolecules between the nucleus and cytoplasm. It is a large, highly organized protein complex consisting of multiple copies of ~30 different nucleoporin proteins. Although the majority of these nucleoporin molecules exist within this complex, a subset have additional roles within the nucleus or cytoplasm as individual proteins. While many studies in the field using artificial overexpression disease model systems have reported cytoplasmic accumulations of a small subset of these nucleoporins, there exists a critical need to evaluate whether or not these proteins are altered within the NPC structure itself. Using super resolution microcopy and human induced pluripotent stem cell-derived neurons, my work has now addressed this gap in knowledge in human neurons from patients with the most common genetic form of ALS. Moreover, I have identified the nucleoporin POM121 as the critical initiator of a pathological cascade impacting NPC composition, function and downstream neuronal survival. Collectively, my work advances our knowledge of human neuronal NPC biology in health and disease and identifies compositional changes to the NPC as an early event in ALS pathogenesis.

Macrophages are a class of innate immune cells that play essential roles in the progression of a variety of major human diseases. My research project is to build multiscale computational models to mechanistically simulate and investigate the role of macrophages, especially their phenotypic polarization, in the regulation of blood vessel formation, inflammation and immune response in disease settings such as cancer and peripheral arterial disease. These data-driven computational platforms that I built were used to identify and evaluate novel therapeutic strategies, with the potential to improve disease outcomes in patients. My research adviser is Aleksander S. Popel, Ph.D., in the Department of Biomedical Engineering.

I completed my Ph.D. thesis work with Kathleen H. Burns, M.D., Ph.D. My work centers around a basic question: How do cells survive the expression of LINE-1 retrotransposons? We believe this question is important in understanding cancer biology, since it seems that cancers, broadly speaking, tolerate reactivation of these transposons. The curious bit was that “normal” cells have a lot of trouble growing while also expressing these transposons, so that clued us in that there must be some way cancer-like cells handle that toxicity. We think that understanding these mechanisms could reveal something fundamental about the selective pressures imposed on tumors during their evolution, and could inform therapy. At the time we started, few in the LINE-1 field were interested in the translational potential for this biology, and even fewer cancer labs were really excited by transposons. We saw an opportunity to bridge this gap.

In the lab of Patricia Janak, Ph.D., I studied how neurons in the ventral pallidum, a brain region considered part of the basal ganglia, respond when rats receive highly palatable food rewards. Historically, the ventral pallidum has been described as a relay station for reward-related information leaving the nucleus accumbens, a more widely studied region known to be important for controlling the seeking and consumption of rewards. In my research, I found that the ventral pallidum has very pronounced responses to rewards that reflect the rats’ relative preference for each available reward. These responses were stronger than in nucleus accumbens and also preceded the similar activity there. This finding contributes to building evidence that the ventral pallidum is important for processing rewards above and beyond its role as an output of the nucleus accumbens. I further found that the neural activity in the ventral pallidum tracks rats’ preferences as they change over time, and this activity guides their choice behavior. These findings should increase interest in studying the ventral pallidum in order to understand how our brain processes information about the rewards we consume and how that information impacts what rewards we choose to consume in the future.

I worked in the lab of Mary Armanios, M.D., where we discovered a new genetic cause of familial pulmonary fibrosis due to a mutation in the nuclear RNA exosome targeting component ZCCHC8. Our lab efforts are focused on elucidating the genetic underpinnings and mechanisms of lung disease, cancer and telomere syndromes. Specifically, my dissertation focused on identifying new genetic causes of pulmonary fibrosis, which affects upward of 100,000 people, with over 30,000 deaths in the U.S. annually. For this work, we initially found that affected individuals in a family with autosomal dominant pulmonary fibrosis had both short telomeres and low levels of the telomerase RNA component. Using cells from these patients, and a CRISPR knockout mouse model, we found that ZCCHC8 is required for telomerase function and telomere length maintenance. In contrast to the ZCCHC8 partial loss found in patients with pulmonary fibrosis, we also found that complete loss of ZCCHC8 causes pervasive RNA dysregulation with misprocessing of other low-abundance RNAs including cilia components. In vivo, this biallelic loss manifests as neurodevelopmental disease. Our discovery identifies novel regulators of telomerase function that are mutated in disease and underscores the important role telomerase dysfunction plays in pulmonary fibrosis susceptibility. It also provides a model for studying the role of RNA misprocessing in the susceptibility to neurodevelopmental disease.

Human DNA is packaged into the protein-DNA complex called chromatin. The minimal organizational unit of chromatin is the nucleosome, which is composed of a core of histone proteins wrapped by DNA. Cells decide which genes are turned on and off, using a complex network of covalent modifications on histone proteins. Inappropriate attachment of histone modifications can cause genes to be turned on when they should be turned off instead, which can lead to a variety of human diseases, such as cancer. For my research, in the lab of Dr. Cynthia Wolberger, I determined the three-dimensional structures of two important enzymes, called Dot1L and COMPASS, bound to nucleosomes. Both Dot1L and COMPASS attach methyl groups to histones, and dysregulation of these enzymes can lead to cancer in humans. My structures revealed how Dot1L and COMPASS interact with the nucleosome to attach methyl groups and how these enzymes are activated by another histone mark, called ubiquitin. My findings answered long-standing questions in the chromatin field about how these enzymes are regulated and opened up new avenues for drug discovery that are currently being pursued in the Wolberger lab.

As an M.D./Ph.D. student in Hongjun Song’s Lab, I had the opportunity to collaborate with neurosurgeons and pathologists to develop a three-dimensional organoid model for glioblastoma directly from surgically resected tumor tissue. By avoiding cell dissociation and clonal selection, glioblastoma organoids better recapitulate the cellular, transcriptional and mutational diversity of patients’ tumors, providing a valuable tool to test targeted therapeutics. My hope is that this model system will be utilized to test chemotherapeutic drugs and immunotherapies targeted to each patient’s unique tumor before initiating treatment, to maximize the probability of survival for those affected by this devastating disease.

I conducted this research in the laboratory of Dr. Sergi Regot. In the Regot Lab, we use biosensors and live-cell imaging to study a wide range of complex biological processes, from embryo development to immune signaling. My recent work seeks to understand innate immunity in the context of the epithelial monolayer, the outermost layer of cells that is often in close proximity to microbes. Epithelial tissues must quantify these microbes in order to maintain a balanced immune state. To understand how this quantification is achieved, we designed a screen of microbe-related signals and measured the nuclear translocation and transcriptional activation of the master immune regulator —NF-κB — in thousands of single cells. We found that as a population, epithelial cells respond to bacterial molecules by activating fractions of cells in an all-or-nothing (i.e., digital) manner. An epigenetic switch that regulates receptor expression maintains the responding cell fraction. The fraction of responding cells can be increased by prolonged innate immune signaling via cytokines that trigger oscillatory NF-κB activity. Epigenetic variegation of receptor expression in single cells allows for long-term, quantitative flexibility in the upper limit of response in the population. This flexible variability at the single-cell level could contribute to the development of diseases that involve chronic inflammatory states tissues.

I performed my thesis research in the laboratory of Xinzhong Dong, Ph.D. With Xinzhong’s guidance, I discovered that bile activates sensory neurons through a receptor called MRGPRX4. This is important because patients who have difficulty excreting bile, such as those with liver damage, experience a profound, whole body itch which is currently untreatable. Our discovery provides better understanding of why this itch occurs and, importantly, identifies a promising therapeutic target to benefit this long-suffering patient population.

Systemic sclerosis (SSc) is a complex disease in which previously healthy young adults show an intractable predisposition for fibrosis of the skin and internal organs. It is among the most devastating rheumatic disease, with a 10-year mortality rate approximating 30%. In the absence of a major defined genetic contribution to disease, the pathogenic mechanism remains unknown. Treatments are largely symptomatic and often ineffective. Under the mentorship of Hal Dietz, M.D., I found that dermal fibroblasts in SSc patients maintained elevated expression of a dominant profibrotic cytokine called TGFB2, through constitutive epigenetic activation of a novel enhancer. This pathologic epigenetic mechanism was found highly dependent on another player called BRD4, the pharmacologic inhibition of which resulted in reversal of dermal fibrosis in SSc skin explants. These findings advance an epigenetic mechanism of fibrosis in SSc and inform a regulatory mechanism of TGFB2.

Local acidosis causes tissue damage, and pain and is one of the hallmarks of ischemia, cancer and inflammation. However, the molecular mechanisms of the cellular response to acid are not fully understood. Besides the cation-conducting, acid-sensing ion channels (ASICs), the acid also activates a chloride (Cl−) conductance in a wide range of mammalian cells. Although the electrophysiological properties of the proton-activated Cl− channel have been described in detail, its molecular identity has remained elusive. This gap makes it impossible to elucidate its precise biological function and potential contributions to the pathogenesis of acidosis-related diseases. Taking advantage of the powerful functional genomics screen established in our lab, we identified a novel membrane protein, named PAC (also known as TMEM206), as the proton-activated Cl− channel. PAC mRNA is expressed in diverse tissues, with the highest level detected in the brain. Knockout of mouse PAC abolishes the channel activity in neurons and protects them from acid-induced cell death. Importantly, PAC KO mice exhibited significantly smaller brain infarct volume when subjected to a middle cerebral artery occlusion (MCAO) stroke model. Therefore, PAC may represent a potential drug target for stroke and other acidosis-associated diseases. This work has been done in the laboratory of Zhaozhu Qiu, Ph.D., in the Department of Physiology.

Navigation is an essential behavior for the survival of a wide range of animal species. The brain must execute fast and efficient ‘where-to-go’ decisions towards a goal to achieve desired outcomes. My research, conducted in Hyung-Bae Kwon’s lab, has shown the cellular and circuit underpinnings of cognition underlying “where-to-go” decision. Particularly, we focused on neural representations of a spatial goal, thought to guide map-based navigation towards the goal. To examine a goal system in the brain, we targeted the nucleus accumbens (NAc), a locus of receiving inputs about spatial-information, reward, and of interfacing outputs for action control. We identified a ‘functionally-defined’ NAc neuronal population that encodes a goal location in the context of valued environment through the medium of dopamine and that truly guides navigation towards the goal by directly manipulating them. Finding goal-directing neurons in the NAc extends the concept of the cognitive map beyond hippocampal-entorhinal circuits and offers advances in understanding fundamental operating mechanisms of flexible action control to pursue diverse behavioral goals stored in cognitive maps.

I began studying how cells control the quality of the messenger RNA they produce in the laboratory of Dr. Rachel Green. We set out to find novel factors associated with the decay of translationally problematic mRNAs and performed genetic screens in collaboration with Dr. Grant Brown’s lab at the University of Toronto. We identified a key new protein that enzymatically cleaves problematic mRNAs at stalled ribosomes. This finding expands our knowledge of how cells respond to errors in transcription and RNA damage.

In the lab of Dr. Alexis Battle, we analyzed genetic data from over 400,000 individuals and discovered a new area of the genome that is associated with the development of clinical heart failure. We then confirmed that finding in an independent cohort of over a million participants. The novel DNA region had a previously unknown function, so we employed several computational algorithms to combine a series of other genetic data (gene expression, epigenetic regulation and 3D-DNA folding information) and predicted that the novel region affects the expression of a nearby structural gene, known as ACTN2. We then confirmed that prediction with direct genome editing, using CRISPR/Cas9 in differentiated cardiomyocytes. The discovery is important, as it increases our understanding of heart failure pathogenesis and could, in the future, be used to identify people at risk or to develop novel therapies for the disease.

In our nose, the sense of smell is mediated by the olfactory mucosa, a neuroepithelial tissue containing long-lived olfactory stem cells. Even though these stem cells possess a remarkable capacity for regeneration, the loss of the sense of smell is a common symptom in the setting of chronic inflammatory rhinosinusitis. Deepening our understanding of the stem cell behavior has the potential to inform novel treatment strategies for inflammation-related olfactory deficits. In Andrew Lane’s Lab, my research focuses on the cross-talk between olfactory stem cells and the local immune system in acute or chronic inflammation. We observed that the basal stem cell population was activated in the inflammatory environment and directly contributed to the disease pathology. Our study establishes a mechanism of chronic rhinosinusitis-associated olfactory loss, caused by a functional switch of neuroepithelial stem cells from regeneration to immune defense. The immune activity of basal stem cells in communicating with infiltrating inflammatory cells may play a more generalized role in mucosal immunity at epithelial barrier surfaces in health and disease.

Type 1 diabetes (T1D) is a major childhood autoimmune disease that results from the destruction of insulin-producing beta cells by islet-reactive T cells, making patients on dependent on insulin replacement for survival. However, insulin replacement is not a cure and many patients suffer long term complications, including cardiovascular, renal and neuropathy. We hypothesize that lack of therapy to protect at risk individuals or slowing loss of beta cells in newly diagnosed patients are a result of lack of key information regarding how the disease develop. I am doing my research at Hamad’s lab in the department of Pathology. We have discovered a new adaptive immune cell that combines lineage characteristics of both B and T cells and clonally expanded in Type 1 Diabetes patients. We refer to this hybrid lymphocytes as dual expressers (DEs) because they co-express the B cell receptor (BCR) and the T cell receptor (TCR) and we generally call them “X cells” to denote their crossover phenotype. Phenotypic and functional characteristics of X cells which are recently published in the prestigious Journal Cell are expected to open a new line of research that can lead to new breakthroughs in the field of autoimmunity that are relevant not only to type 1 diabetes but also are other autoimmune diseases such as multiple sclerosis, lupus. Briefly, our findings indicate that the X cell are major drivers of type 1 diabetes by bearing an insulin mimic that cross-stimulates insulin-reactive T cells that go on to infiltrate pancreas and destroy insulin-producing beta cells. We believe these findings together with our ongoing research are preparing the platform for developing a biomarker that helps screen individuals at risk for developing Type 1 diabetes at very early age, possibly at birth and lay the groundwork for developing immunotherapies that target X cells for elimination for protecting at risk individuals.

Every movement begins and ends in a period of stillness. In the Laboratory for Computational Motor Control (the Shadmehr Lab), we study how the brain controls these different periods of motor activity. While decades of research have demonstrated that one area of the brain, the primary motor cortex, is critical for the execution of a movement, we know comparatively little about how the brain holds the arm still in a desired posture. In our recent work, we wondered if the brain holds the arm still using a strategy similar to that of the eye. For the eye, there are some neurons that produce a “moving” signal that corresponds to the velocity of the eye. However, there is a separate set of neurons that produce a “holding” signal that corresponds to the position of the eye. Much like the formula “distance equals rate multiplied by time,” the holding neurons calculate the “holding” signal by mathematically integrating the “moving” signal over time. Through a sequence of experiments involving over 200 healthy humans, 14 stroke patients, and four non-human primates, we discovered that a very similar integration process is used to hold the arm still. Critically, because reach integration was unimpaired in patients who suffered from cortical strokes, our work suggests that there are separate areas of the brain that move the arm and hold the arm still, as for the eye. These findings may help us understand why some neurological conditions can lead to impairment in movements and abnormal postures.

Sound detected by a single auditory hair cell is relayed by 10-30 type I spiral ganglion neurons (SGNs), the primary sensory neurons of the auditory system, to the central nervous system. Although these neurons exhibit differences in spontaneous firing rates and activation thresholds, it is unclear whether the physiological diversity reflects their endogenous heterogeneities and whether they differentially process auditory information. In Ulrich Mueller’s lab, I demonstrated that type I SGNs consist of three molecularly distinct subtypes (named type IA, type IB and type IC), using single-cell RNA sequencing (scRNAseq). These three subtypes exhibit distinct innervation patterns and spontaneous firing rates. Moreover, these subtypes are specified postnatally in an activity-dependent manner. Deafness mutations that disrupt hair cell mechanotransduction and glutamatergic communications from hair cells to SGNs disrupt this SGN subtype specification. To further characterize the functional differences of SGN subtypes, I have generated CreERT2 mouse lines that specifically target the type IA and IC populations. Based on their physiological properties, the low spontaneous firing neurons with high activation thresholds are predicted to be dispensable for hearing sensitivity, but their large dynamic ranges are expected to play critical roles in hearing in noise. Consistently, silencing the type IC subtype significantly impaired hearing in a noisy environment, but not in quiet. Surprisingly, inhibition of the type IC population also affected frequency discrimination, suggesting previously unappreciated fundamental functional differences among SGN subtypes.

I work under the mentorship of Drew Pardoll, M.D., Ph.D.; Janis Taube, M.D., M.Sc.; and Alex Szalay, Ph.D.; at the exciting interface of cancer immunology, pathology and data science. The team’s long-term goal is to create carefully curated, open source atlases of different tumors stained for multiple proteins, thus spatially mapping immune activity within the tumor microenvironment. Similar to The Cancer Genome Atlas and Protein Data Bank, we hope that making large data sets along with specialized data analysis tools publicly available will spur discoveries at an unprecedented level in cancer immunotherapy. My thesis work has formed the foundation of this effort. Specifically, I developed an end-to-end platform with meticulous quality control for creating quantitative, spatially resolved multiplex immunofluorescence data using lessons derived from the field of astronomy. Implementation of this platform led to the identification of two previously under-recognized immunoactive cell subsets that are now important biomarker candidates for defining survival after anti-PD-1 therapy in patients with melanoma.

Unidirectional energy-dependent drug efflux mediated by cellular membrane proteins results in the failure of many anti-cancer chemotherapeutic agents. One strategy to enhance tumor retention of imaging agents, or anti-cancer drugs, is designing probes that undergo a tumor-specific enzymatic reaction that prevents them from being pumped out of the cell. In the Bulte lab, we used a rational design approach to develop a cell-penetrating, small molecule probe — Olsa-RVRR — which consists of the anti-cancer agent olsalazine (Olsa), conjugated to the furin-targeted peptide RVRR. By virtue of a biologically compatible condensation reaction, Olsa-RVRR monomers can be subjected to furin-induced intracellular condensation to form CEST MRI-detectable and therapeutic nanoparticles (Olsa-NPs) in targeted tumor cells. Hence, our work provides a potential platform for imaging tumor aggressiveness, drug accumulation and therapeutic response.

In King-Wai Yau’s lab, I study the phototransduction mechanisms of intrinsically photosensitive retinal ganglion cells (ipRGCs). Photoreceptors generally have a specialized photosensitive compartment derived from a modified cilium or microvilli, and are classified as being either rhabdomeric (microvillous) or ciliary. The corresponding phototransduction mechanism, although often varied in details, follows the morphology and is either PLC- or cyclic-nucleotide-mediated. Unlike most photoreceptors, ipRGCs show no signs of a microvillous (rhabdomeric) or ciliary photosensitive compartment. Previous work from our lab showed a PLC-mediated pathway in M1-ipRGCs. The question remains whether other ipRGC subtypes have the same phototransduction mechanism. My colleague and I revealed a new phototransduction pathway in M2- and M4-ipRGCs mediated by cyclic nucleotide and HCN channels. The coexistence of both PLC- and cyclic-nucleotide-mediated pathways in a given M2-ipRGC breaks a dogma about the general segregation of the two phototransduction motifs in different cells, although a copresence has been postulated previously by some developmental biologists to exist at the beginning of photoreceptor evolution. More interestingly, the specific second messenger is cAMP, instead of being cGMP as found for nearly all ciliary phototransduction mechanisms across the animal kingdom except jellyfish (an evolutionarily ancient species). These findings suggest ipRGCs are an extant ancestral photoreceptor.